William Spagnola

@williamspagnola

I tweet/retweet about stats

New York, NY
Vrijeme pridruživanja: svibanj 2019.

Tweetovi

Blokirali ste korisnika/cu @williamspagnola

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @williamspagnola

  1. proslijedio/la je Tweet
    31. sij

    Slides from my talk about queryparser, tidyquery, and all things SQL ⇄ dplyr! I'll share the recording when it's up

    Poništi
  2. proslijedio/la je Tweet
    31. sij

    The way we talk about data science and focus so much on methods, we actually incentivize working with *bad* data, rather than spending the time to collect good data and then use easy methods with it

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet

    Really good point. An instance of people hastily picking up on the latest trendy thing.

    Poništi
  4. proslijedio/la je Tweet
    31. sij

    "What did you do as a new T that you'd NEVER do now?" Work in a school: 1. without a clear and coherent culture; 2. that had no curriculum; 3. wasn't committed to knowledge-rich literacy; and 4. doesn't understand that low expectations applies to academics AND behavior.

    Poništi
  5. proslijedio/la je Tweet
    28. sij

    Hope everybody is having a good week so far. May all your datasets be clean and all your p's be less than 2.22*10^−16!

    Poništi
  6. 28. sij

    Subway delays are inevitable even in a good system but there is no excuse for not fixing PA system

    Poništi
  7. 27. sij

    This is an event that really should get more press

    Poništi
  8. proslijedio/la je Tweet
    21. sij

    More formally, the study made claims about P(shot|race), but estimated P(race|shot). Those are *not* the same. That’s not a statement of preference. It’s a mathematical fact, Bayes’ rule, that we’ve known since the 1700s. 8/n

    Prikaži ovu nit
    Poništi
  9. 15. sij

    It does seem that most people who talk about how much easier python is than R come from a CS background

    Poništi
  10. proslijedio/la je Tweet
    15. sij

    Ya no estoy seguro de que Python sea un buen lenguaje para iniciarse en Data Science. En los últimos años, trabajando con estudiantes con potencial pero sin una formación basada en Computer Science, me he dado cuenta que Python exige un modelo mental de CS para poder avanzar.

    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    11. sij

    I still love this slide so much 😂

    DSLearnings: Avoid Analysis Mistakes (Shows suspects police lineup) understanding data sources; assumptions; filters; missing values; understanding of model goals
    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    10. sij

    Adam 's talk on Bayesian Factorization Machines is now available: . The slides are also up: . Have a nice weekend!

    Poništi
  13. proslijedio/la je Tweet
    10. sij

    Bayesian factorization in both and

    Poništi
  14. proslijedio/la je Tweet
    9. sij

    There are moments with Python that feel like I'm writing with God's pseudocode itself. R is pseudocode according to the new testament of statistics. R embodies statistical thinking as code. While Python feels right to the inner computer scientist, R appeals to inner statistician.

    Poništi
  15. proslijedio/la je Tweet
    9. sij

    Today in my dept's journal club: "10 Simple Rules for Computational Modeling of Behavioral Data" I've been making my PhD students simulate their projects prior to data collection. This paper looks like a nice guide they could have used 6 months ago...

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    8. sij

    twitter, I like you better than the rest of twitter. I move to secede. Anyone second?

    Poništi
  17. proslijedio/la je Tweet
    8. sij

    If 1024 fair coins are each tossed 10 times, chances are good (> 63%) that at least one will come up heads 10 times in a row; and that coin will be proud to explain how its skill, faith, guts & determination made its achievement possible, and how that combo can work for you too.

    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet
    7. sij

    At the end of 2020, instead of 30 Under 30 and NextGen lists, please profile middle-aged people who just got their big breaks. I want to read about a mother of 2 who published her first novel, a director who released their first studio feature at 47, THAT'S THE LIST WE WANT.

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    7. sij

    🤖 Some great resources, including 's ML for Everyone which is full of fun drawings! 👩‍🏫 "Learning to Teach Machines to Learn" by

    Classical Machine Learning tree by @vas3kcom source: https://vas3k.com/blog/machine_learning/#scroll50
    Poništi
  20. proslijedio/la je Tweet
    6. sij

    My new coffee mug is also a git cheat sheet!

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·